Apatinib inhibits the growth of small cell lung cancer via a mechanism mediated by VEGF, PI3K/Akt and Ki‐67/CD31

Abstract This study aimed to investigate the anti‐tumour effect of apatinib on extensive‐stage small cell lung cancer (SCLC) and elucidate the associated mechanisms. NCI‐H345 cells were selected as model cells because of high expression of vascular endothelial growth factor (VEGF), VEGF receptor 2 (VEGFR2) and phosphorylated‐VEGFR2 (pVEGFR2). Cells were exposed to recombinant human VEGF (rhVEGF) and apatinib. Cells were then divided into eight groups, namely, control, rhVEGF, apatinib, rhVEGF+apatinib, serum‐free medium (SM), SM+rhVEGF, SM+apatinib and SM+rhVEGF+apatinib. In comparison with the control group, cell proliferation in vitro in apatinib, SM, SM+apatinib and SM+rhVEGF+apatinib groups was inhibited, particularly in SM+apatinib group. The effect of apatinib on tumour growth in vivo was investigated using a mouse xenograft tumour model. In comparison with the control group, tumour sizes were reduced in apatinib‐treated group on days 34 and 37. Immunohistochemical and immunofluorescence staining revealed that VEGF, pVEGFR2, PI3K, AKT, p‐ERK1/2, Ki‐67 and CD31 in the tumour cells of apatinib‐treated group were downregulated compared with control group. Haematoxylin and eosin staining revealed that apatinib promoted the necrosis of SCLC cells in vivo. In conclusion, apatinib inhibited the growth of SCLC cells by downregulating the expression of VEGF, pVEGFR2, p‐PI3K, p‐AKT, p‐ERK1/2, Ki‐67 and CD31.

platinum-based first-line chemotherapy, and the efficiency of current second-line agents is poor. [2][3][4] Recent studies have reported novel therapeutic agents such as pegfilgrastim and irinotecan for ES-SCLC. [5][6][7] SCLC tumours are characterized with high microvessel density and a rich blood supply. 8 Vascular endothelial growth factor (VEGF) is upregulated in the serum of the majority of patients with SCLC and is negatively correlated with the sensitivity of chemotherapy and survival time. 8 In the past decades, VEGF and the VEGF receptor (VEGFR) have served important roles in tumorigenesis. [9][10][11] The suppression of VEGF expression by VEGF neutralizing antibody inhibits tumour growth and metastasis, 8 and an increasing number of drugs, such as bevacizumab and sunitinib, which target VEGF and VEGFR, have been developed. [9][10][11][12] Apatinib is a highly selective VEGFR2 inhibitor, which inhibits the angiogenesis of nasopharyngeal carcinoma, non-Hodgkin lymphoma and small-cell carcinoma of oesophagus. [13][14][15][16] The clinical efficacy of apatinib in advanced tumours, including relapsed or refractory non-Hodgkin lymphoma, metastatic gastric cancer, colorectal cancer, non-SCLC (NSCLC), neuroendocrine tumours and mesothelioma, has been demonstrated in several clinical trials. 14,15 Therefore, apatinib may be a suitable therapeutic agent for the maintenance treatment of patients with extensive-stage (ES)-SCLC. However, to the best of the authors' knowledge, the efficacy of apatinib as maintenance therapy in ES-SCLC has not been reported, and the efficacy, safety and mechanisms of apatinib monotherapy in ES-SCLC deserve further investigation. Accordingly, the present study aimed to investigate the anti-tumour effect of apatinib in ES-SCLC by using in vivo and in vitro models.

| Cell lines, animals and apatinib
The human SCLC cell lines (NCI-H345 and NCI-H446) were obtained from BeNa Culture Collection (BNCC

| Cell screening and rhVEGF treatment
The levels of VEGF, VEGFR2, and pVEGFR2 in NCI-H345 and NCI-H446 cells were examined by Western blot analysis to select the   ideal cell model. NCI-H345 and NCI-H446 cells were subsequently   treated with 30 ng/ml recombinant human VEGF (rhVEGF; cat. no 10542-H08H; Sino Biological) for 0, 30, 60 and 120 min at 37°C.
The levels of VEGFR2 and pVEGFR2 in the cells were examined by Western blot analysis to identify the optimal exposure time.

| Western blot analysis
Cells were lysed in radioimmunoprecipitation assay buffer (Beyotime) at 4°C for 30 min and centrifuged for 10 min at 4°C and 9000 ×g. Total protein concentration was quantified using bicinchoninic acid assay. Proteins were subsequently denatured by boil-

| Cell proliferation in vitro
NCI-H345 cells were seeded at a density of 10,000 cells per well in a 96-well plate and cultured at 37°C and 5% CO 2 for 24 h. The cells were treated with increasing concentrations of apatinib (50, 100, 150, 200 and 250 nM) for different durations (24,48 and 72 h). The concentration of apatinib was selected as previously described (17).
A total of 10 μL of cell counting kit-8 (CCK-8) solution were added per well, and the cells were incubated for 2 h. Cell viability (%) was calculated by measuring the absorbance at a wavelength of 570 nm by using a microplate reader (cat. no. RT-6100; Rayto).
In addition, the cells were divided into eight groups as follows: (i) control (without any treatment); (ii) rhVEGF (cells were treated with 30 ng/ml rhVEGF for 60 min); (iii) apatinib (cells were treated with 250 nM apatinib for 60 min); (iv) rhVEGF+apatinib (cells were treated with 30 ng/ml rhVEGF and 250 nM apatinib for 60 min); (v) serum-free medium (cells were cultured in serum-free medium for 12 h); (vi) serum-free medium+rhVEGF (cells were cultured in serumfree medium for 12 h and then treated with 30 ng/ml rhVEGF for 60 min); (vii) serum-free medium+apatinib (cells were cultured in serum-free medium for 12 h and then treated with 250 nM apatinib for 60 min); and (viii) serum-free medium+rhVEGF+apatinib (cells were cultured in serum-free medium for 12 h and then treated with 30 ng/ml rhVEGF and 250 nM apatinib for 60 min). Cell proliferation was investigated in each group via CCK-8 assay.

| Tumour growth in vivo
Eight nude mice were anaesthetized by inhaling ether. A total of 1 × 10 7 NCI-H345 cells in 0.2 ml of normal saline were injected subcutaneously into the right axillary region on day 0. Cerebral state index, bunt suppression, electromyographic and signal quality index were monitored to ensure that the animals were anaesthetized rather than euthanized after exposure to ether. Tumour sizes were calculated approximately twice a week by using the following equa-

| Immunohistochemical analysis
Tumour tissues were fixed in 4% paraformaldehyde at room temperature for 10 min, embedded in paraffin and cut into sections (4 μm thickness). The sections were dehydrated in 70%, 80% and 90% ethanol solutions and dewaxed in xylene twice and rehydrated.

| Immunofluorescence analysis
The paraffin sections were dehydrated, dewaxed and rehydrated. Semi-quantitative analysis was performed using ImageJ software (version 1.8.0; National Institute of Health).

| Statistical analysis
Data were presented as mean ± standard deviation. Statistical analysis was performed using SPSS software (version 19.0; IBM Corp.).
One-way analysis of variance and Tukey post hoc test were used to compare the different groups. Significant difference was considered at p < 0.05.

| Cell line screening and rhVEGF exposure time
The levels of VEGF, VEGFR2 and pVEGFR2 in NCI-H345 and NCI-H446 cells were assessed by Western blot analysis (Figure 1).
Although the ratio of pVEGFR2 versus VEGFR2 was similar between the two cells, compared with NCI-H345 cells, the expression of VEGF, VEGFR2 and pVEGFR2 in NCI-H446 cells was significantly reduced (p < 0.05). Therefore, NCI-H345 cells were selected for experimentation. The levels of VEGFR2 and pVEGFR2 in NCI-H345 cells following treatment with rhVEGF for 0, 30, 60 and 120 min were investigated by Western blot analysis ( Figure 2). The ratio of pVEGFR2 versus VEGFR2 was similar among various treatments.
However, the expression levels of VEGFR2 and pVEGFR2 in rhVEGFtreated NCI-H345 cells were significantly upregulated compared with the controls (p < 0.05). Considering that the expression levels of VEGFR2 and pVEGFR2 were the highest at 60 min, this time point was selected for experimentation.

| Cell proliferation in vitro
The proliferation rate of NCI-H345 cells treated with different con- Therefore, the highest concentration of apatinib (250 nM) was selected for subsequent experimentation. However, considering that a long exposure time may reduce cell proliferation and affect the results, an exposure time of 60 min was selected.
The proliferation rate of NCI-H345 cells in the eight groups was evaluated by the CCK-8 assay ( Figure 3B). The cells in the control, rhVEGF, rhVEGF+apatinib and serum-free medium+rhVEGF groups exhibited similar proliferation rates (p > 0.05). In comparison with the control group, the proliferation rate of cells in the apatinib, serum-free medium, serum-free medium+apatinib and serum-free medium+rhVEGF+apatinib groups significantly decreased, particularly in the serum-free medium+apatinib group (p < 0.05).

| Tumour growth in vivo
Tumour growth in mice following apatinib treatment was investi-

| Expression of VEGF, pVEGFR2, PI3K, p-PI3K, AKT, p-AKT, ERK1/2, p-ERK1/2 and Ki-67
The expression level of VEGF, pVEGFR2, PI3K, p-PI3K, AKT, p-AKT, ERK1/2, p-ERK1/2 and Ki-67 in tumour tissues harvested from the xenograft model was determined by immunohistochemical analysis F I G U R E 1 Levels of VEGF, VEGFR2 and pVEGFR2 in NCI-H345 and NCI-H446 cells which were examined by Western blotting. All blots were normalized to GAPDH levels. * p < 0.05 vs. NCI-H345 cells. VEGF, vascular endothelial growth factor; VEGFR2, vascular endothelial growth factor receptor 3; p, phosphorylated ( Figure 5). In comparison with the mice in the control group, the expression levels of VEGF, pVEGFR2, p-PI3K, p-AKT, p-ERK1/2 and Ki-67 in tumours obtained from apatinib-treated mice were significantly downregulated (p < 0.05). Negative control staining by using an irrelevant isotype antibody was performed to ensure that the staining for the markers of interest was specific.

| CD31 expression
The expression level of CD31 in tumour tissues harvested from the xenograft model was determined by immunofluorescence analysis ( Figure 6). In comparison with the mice in the control group, the expression level of CD31 in tumours obtained from apatinib-treated mice was significantly downregulated (p < 0.05). Negative control staining by using an irrelevant isotype antibody was performed to ensure that the staining for CD31 was specific.

| Haematoxylin and eosin staining
Representative haematoxylin and eosin staining images of the tumour tissues obtained from apatinib-treated mice are presented in Figure 7. The control group exhibited a uniform distribution of F I G U R E 2 Levels of VEGFR2 and pVEGFR2 in NCI-H345 cells following treatment with treatment with rhVEGF for 0, 30, 60 and 120 min analysed by Western blotting. All blots were normalized to GAPDH levels.

| DISCUSS ION
Apatinib mesylate is a small molecule VEGFR2 tyrosine kinase inhibitor developed in China that can be administered orally. Apatinib competitively binds to the adenosine triphosphate binding site in VEGFR2, thus blocking the downstream signalling pathways and inhibiting tumour angiogenesis. [18][19][20][21] VEGF has an angiogenesispromoting effect. 22 Additionally, the VEGF autocrine signalling pathway in tumour cells plays an important role in tumour growth. 22 Tumour cells in different types of cancer, including cholangiocarcinoma, hepatocellular carcinoma and NSCLC, 23-25 co-express VEGF and VEGFR2, and the interaction between these two proteins promotes the proliferation of tumour cells. 26 The present study revealed that apatinib inhibited SCLC cell proliferation in vitro, and this effect was inhibited by exogenous rhVEGF.
In addition, tumour sizes in apatinib-treated mice apatinib were smaller than those of control mice. Therefore, apatinib inhibited the growth of SCLC tumours in vitro and in vivo. The tumour volume on day 37 decreased possibly because the effect of apatinib on the tumour became increasingly obvious with prolonged administration time. Additionally, the expression levels of VEGF and pVEGFR2 in tumour tissues were downregulated following apatinib treatment, suggesting that apatinib inhibited the expression of VEGF and pVEGFR2 in SCLC tumours.
The PI3K/AKT signalling pathway is involved in tumorigenesis and activates the expression of downstream genes or proteins to induce cell proliferation, regulate downstream apoptotic or anti-apoptotic genes, mediate haematopoiesis and angiogenesis, and increase oxygen utilization by tumour cells by regulating glucose transport and glycolysis-associated enzymes. 27 ERK1/2 is a member of the mitogen-activated protein kinase family. 28 The activation of ERK1/2 signalling pathway is closely associated with the development and progression of various tumours, such as renal cancer, hepatocellular carcinoma, prostatic carcinoma and NSCLC. 29 Apatinib does not affect the expression of total PI3K, AKT and ERK1/2. 17,30 The expression of p-PI3K and p-AKT in SCLC tumours was remarkably decreased following apatinib treatment, suggesting that apatinib may inhibit SCLC tumour growth by inhibiting the PI3K/AKT signalling pathway. The results of the present study revealed that the expression levels of p-ERK1/2, Ki-67 and CD31 in SCLC tumours were significantly reduced following apatinib treatment, suggesting that apatinib may also inhibit SCLC tumour growth by inhibiting the expression of p-ERK1/2, Ki-67 and CD31. ERK1/2 is activated in NSCLC and is associated with advanced tumours. 29 A limitation of the present study is that Western blot experiments were not conducted to determine the levels of the phosphorylated and total proteins mentioned above.

Ki-67 is a nucleoprotein involved in ribosomal RNA transcription
and is expressed at all stages of cell proliferation, except the G0 stage. 31 Ki-67 inactivation inhibits ribosomal RNA synthesis, and Ki-67 is used as a biomarker of cell proliferation. 31 The Ki-67 index in pathological reports is closely associated with the differentiation, invasion, metastasis and prognosis of tumours, such as cervical cancer. 32 Increased Ki-67 expression in tumours is associated with increased degree of malignancy, increased tendency for invasion and metastasis, and poor prognosis. 33 CD31 is mainly used to identify endothelial cells and evaluate tumour angiogenesis, which may indicate the rate of tumour growth. 34 Malignant vascular endothelial cells generally retain antigens, thus allowing the use of CD31 expression to diagnose haemangioma and angiosarcoma. 34 The present study mainly aimed to determine the anti-tumour effect of apatinib on ES-SCLC, but the research on related mechanisms is not systematic enough. Only the effects on some pathways or indicators related to tumour are revealed. We will F I G U R E 4 Tumour size-time curve exhibiting tumour growth in a tumour xenograft model in apatinib-treated and control mice. * p < 0.05 vs. control F I G U R E 5 Expression of VEGF, pVEGFR2, PI3K, p-PI3K, AKT, p-AKT, ERK1/2, p-ERK1/2 and Ki-67 in tumour tissues was determined by immunohistochemical analysis. Negative control staining was performed using irrelevant isotype antibody. The control images were obtained from the cells/tissues of control animals. The control staining image was the negative control, which was used to exclude the non-specific reaction of the antibody. * p < 0.05 vs. control. VEGF, vascular endothelial growth factor, p, phosphorylated; VEGFR2, vascular endothelial growth factor receptor 2; PI3K, phosphoinositide 3-kinase; AKT, protein kinase B; Ki-67, marker of proliferation Ki-67 further explore the effects of apatinib on various solid tumours and determine the effect of apatinib on the PI3K/AKT signalling pathway.
In conclusion, the present study revealed that apatinib inhibited SCLC tumour growth in vitro and in vivo by downregulating the expression of VEGF, pVEGFR2, p-PI3K, p-AKT, p-ERK1/2, Ki-67 and CD31. These results suggest that the potential application of apatinib for the treatment of ES-SCLC requires further investigation and may aid in the identification of novel therapeutic targets and diagnostic markers.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Ning Zhong: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); F I G U R E 6 CD31 expression in tumour tissues was analysed by immunofluorescence staining. CD31 protein is presented in red and DAPI in blue. Negative control staining was performed using irrelevant isotype antibody. The control images were obtained from the cells/tissues of control animals. The control staining image was the negative control, which was used to exclude the non-specific reaction of the antibody. * p < 0.05 vs. control. CD31, CD31 antigen

DATA AVA I L A B I L I T Y S TAT E M E N T
The analysed data sets generated during the study are available from the corresponding author on reasonable request.